Bilateral vestibular failure (BVF) due to ototoxic medications, Meniere's disease, effects of genetic syndromes and other insults to the inner ear can cause disabling symptoms. Affected individuals suffer chronic disequilibrium and oscillopsia, and those who fail to compensate for their loss are at increased risk of falls and other injuries. There are currently no adequate treatment options for BVF. Striving to address the needs of these individuals, we have developed a prototype vestibular prosthesis that encodes 3D head movement via 3 gyroscopes and electrically stimulates the corresponding branches of the vestibular nerve. Initial tests of this device in chinchillas have been promising, but further improvement in stimulus encoding will be required to reliably encode the full range of head movements for which the angular vestibulo-ocular reflex normally stabilizes gaze. Trial and error optimization of the ten principal parameters that define stimulus encoding characteristics for each prosthesis channel is impractical. For successful application in a clinical setting, a more efficient approach to stimulus protocol programming will be needed. The first part of the proposed project will establish a robust and efficient approach to optimization of the stimulus parameters, then use data acquired during optimization to refine a model of how the vestibular nerve and central nervous system respond to prosthetic electrical stimuli. While the first human vestibular implant recipients will be individuals with profound bilateral loss, expanding implantation candidacy criteria may lead to implantation of subjects with residual mechanical sensitivity. Understanding whether and how natural and prosthetic inputs are integrated centrally will facilitate design of electrodes and stimulus protocols that make maximal use of residual mechanical sensitivity. At the same time, examining the interplay of responses to natural and prosthetic stimuli offers an opportunity to study the neurophysiology of sensory integration using otherwise impossible experimental paradigms. The second part of the proposed project will characterize the angular vestibulo-ocular reflex (aVOR) in response to electrical and natural stimulation alone and in combination. RELEVANCE: A vestibular prosthesis could restore lost function due to BVF, much as a cochlear implant restores hearing. This project will directly impact the design of stimulus protocols for a vestibular prosthesis for human use, clarify the interaction between prosthetic and residual mechanical sensation in the implanted labyrinth, and refine a "virtual chinchilla labyrinth" computational model which, once validated, can be readily adapted to human anatomy for use in optimal design of electrodes for implantation in the human labyrinth.